Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: : RE: list command


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: : RE: list command
Date   Mon, 4 Aug 2008 17:53:40 +0100

Not quite. 

-egen, count()- does not count distinct values. It counts non-missing
values, regardless of whether they are the same or different. 

See the references in my previous post in this thread. 

Nick 
n.j.cox@durham.ac.uk 

Steven Samuels

I think that Daniel's second example would be:
> Ndc: 00904161260  gpi: 44300040007440
>      00904161260  gpi: 44300040007441

and that he wants to list instances where the same NDC has 2 or more  
GPI's.

If so:

bysort ndc (gpi): gen count=_N   // counts number of observations  
with the same ndc
list  ndc old_ndc e49_ gpi g64_  if count>1

OR
egen count= count(gpi), by(ndc)

list  ndc old_ndc e49_ gpi g64_  if count>1


If the same ndc gpi combination is listed more than once in the data,  
Daniel should first:

bysort gpi ndc: keep if _n==1


On Aug 4, 2008, at 11:51 AM, Daniel Sepulveda-Adams wrote:

> OK here is an example of what happened in the dataset, this dataset  
> is a
> drugs description, therefore codes are associated to names and  
> strength the
> drug and packages size, etc.
>
> Ex:
> 1) ndc (National Drug Code) assigned one gpi (Generic Product  
> Identifier)
>
> ndc: 49502007412   gpi: 44100010102005
>
> 2) ndc that has more than one gpi (9 in this specific case, why  
> because each
> gpi represent a different packages size)
>
> Ndc: 00904161260  gpi: 44300040007440
>
> Please let me know if that clarified my question and thank you for  
> your
> help.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index